Project summary: Ovarian cancer (OC) is a leading cause of cancer deaths in women in the United States. The clinical treatment of OC is marked by high initial response rates to 1st line standard-of-care (SOC) platinum/taxane treatment, but also significant relapse rates. A major unmeet need in this disease exists to identify optimal 2nd line therapy options in the platinum resistant or refractory setting which yield superior outcomes compared with current treatment selection which is largely empirical. Many single and combination agents are currently in use or in clinical trials as 2nd line treatments for OC, but there are few reliable biomarkers available to predict response to these various types of cytotoxic or targeted chemotherapies. Recent studies have identified potential genomic and other markers, including BRCA1/2, VEGF, MAPK, PI3K and others. Advances in biomarker identification, however, are slow because each trial only administers a single therapy on every patient at a given time, and thus only allows biomarker/phenotype correlations on one therapy at a given tumor state. We have developed a technology that allows us to measure the phenotypic effect of a large number of distinct agents or drug combinations within a single tumor in a rapid and minimally invasive manner and without systemic toxicities. The technology represents a new paradigm for predicting proactively, rather than empirically, the effect of drugs inside a patient tumor. Consisting of implantable microdevices, which release microdoses of up to 120 distinct agents or combinations in parallel into small confined regions of tumor, it allows for direct measurement of drug/tumor interaction using established pharmacodynamic readouts for each therapy tested within the native tumor microenvironment. This proposal seeks to translate the microdevices into clinical use through a pilot trial in ovarian cancer patients. Up to 6 microdevices will be implanted into omental tumors one day prior to surgery where they are expected to provide a rapid and comprehensive snapshot of how the tumor actually responds to all the available therapies. 6 spatially separated readouts from distinct parts of the tumor will also provide a novel functional examination of tumor heterogeneity. All of the readouts will be correlated with genomic, transcriptomic, immune and tissue biomarkers, effectively performing 20 or more biomarker trials in each patient with six-fold replicates, with only one extra biopsy procedure, and at minute drug exposure levels. This unprecedented clinical study seeks to validate safety, usage and procedures for obtaining multiple drug phenotypes from microdevices within patients, and to significantly expand the field's understanding of predictive biomarkers in ovarian cancer. If successful, this research may lay the foundation for application of this drug screening microdevice beyond ovarian cancer.